1,2,3,5,3,1- 1Manchester Metropolitan University, Department of Natural Sciences, Manchester, United Kingdom of Great Britain – England, Scotland, Wales (eva.arnau@gmail.com)
- 2Department of Cartographic Engineering, Geodesy and Photogrammetry, Universitat Politècnica de València, Valencia, Spain
- 3Department of Agronomy, University of Almería, Almería, Spain
- 4Department of Cartographic and Land Engineering, Higher Polytechnic School of Avila, University of Salamanca, Ávila, Spain
- 5Centre for Applied Ecology “Prof. Baeta Neves” (CEABN-InBIO), School of Agriculture, University of Lisbon, Lisbon, Portugal
- 6Department of Desertification and Geo-Ecology, Experimental Station of Arid Zones (EEZA-CSIC), Almería, Spain
- 7Soil and Water Conservation Research Group, Centro de Edafología y Biología Aplicada del Segura (CEBAS-CSIC), Murcia, Spain
- 8Department of Geography, University of Valencia, Valencia, Spain
The way vegetation is spatially arranged is a major concern for understanding ecosystem structure and functioning, and it is particularly relevant in drylands due to its control on surface water redistribution. As a result, vegetation spatial configurations derived from Earth Observation (EO) data are widely used as the basis for ecological indicators. However, the transition from pixel-based vegetation mapping to pattern-based interpretation is often implicitly assumed to be straightforward.
In patchy dryland landscapes, where vegetation and bare soil coexist at fine spatial scales, this assumption may be particularly problematic. Spatial structures extracted from EO products are not direct observations of ecological organization, but outcomes of classification and mapping choices operating across multiple spatial scales. Sensor characteristics, spatial resolution and spatial support, together with algorithmic choices, jointly shape how vegetation configurations are inferred and subsequently interpreted.
In this contribution, we examine how different EO-derived estimates of vegetation influence the spatial patterns retrieved in dryland ecosystems. Using aerial and drone imagery across varying spatial and spectral resolutions, we assess the sensitivity of commonly used spatial pattern descriptors to mapping and classification choices, without restricting the analysis to a single methodological approach. Particular attention is given to how pixel-level decisions propagate to landscape-scale pattern characterizations, affecting the apparent configuration of vegetation.
Our results show that spatial pattern metrics vary substantially across EO-derived vegetation products, and that apparent differences in configuration may arise from classification artefacts as much as from genuine ecological structure. This has important implications for the use of spatial patterns as empirical proxies for ecosystem functioning, highlighting the need for more scale-aware mapping workflows and more cautious use of EO-derived spatial patterns in dryland environments.
How to cite: Arnau-Rosalén, E., Marqués-Mateu, Á., Martínez-Sánchez, J. F., Molada-Tébar, A., Rodríguez-Lozano, B., Rodríguez-Caballero, E., Pons-Crespo, R., Lázaro-Suau, R., Castillo-Sánchez, V., Cantón-Castilla, Y., Calvo-Cases, A., and Symeonakis, E.: Patchiness configuration fidelity in dryland vegetation: the EO scale-issue and challenge for spatial pattern surveys, EGU General Assembly 2026, Vienna, Austria, 3–8 May 2026, EGU26-16557, https://doi.org/10.5194/egusphere-egu26-16557, 2026.
